Commits · 7af3bd91dcc035aeeadf469cb68a3137f9e85c21 · ModelZoo / ResNet50_tensorflow

28 May, 2019 5 commits

Merged commit includes the following changes: (#6898) · 7af3bd91

Hongjun Choi authored May 28, 2019

250347237  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Fix linting errors in BERT benchmark test.

--
250326131  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

250315593  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

250303528  by haoyuzhang<haoyuzhang@google.com>:

    Add method docstring to fix lint error.

--

PiperOrigin-RevId: 250347237

7af3bd91

Use more warmup steps for 96 core tests (#6881) · 8b52cd23

Haoyu Zhang authored May 28, 2019

* Run different numbers of steps on different platforms

* Add new tests for delayed performance measurement

8b52cd23

Add shuffle to dataset records · 733a752d
guptapriya authored May 28, 2019
```
This shuffling should help in getting shuffling each epoch.
```
733a752d
Remove extra time_callback · 372ac40a
guptapriya authored May 28, 2019

372ac40a
Fix breakage due to early stopping callback · 719eec7b
guptapriya authored May 28, 2019

719eec7b

26 May, 2019 1 commit

Merged commit includes the following changes: (#6885) · 6fc642d4

Hongjun Choi authored May 25, 2019

250009207  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Add feature in BERT to write training metrics to a summary file.

--

PiperOrigin-RevId: 250009207

6fc642d4

24 May, 2019 7 commits

Merged commit includes the following changes: (#6880) · fa10031d

saberkun authored May 24, 2019

249896208  by hongkuny<hongkuny@google.com>:

    Adds __init__.py

--

PiperOrigin-RevId: 249896208

fa10031d

Add early stopping logic to ncf keras when desired threshold is met. Also... · 7033c8a2

Priya Gupta authored May 23, 2019

Add early stopping logic to ncf keras when desired threshold is met. Also change the default batch size to match the tuned hyperparams

7033c8a2

Merged commit includes the following changes: (#6879) · 7f9db598

saberkun authored May 24, 2019

249883771  by hongkuny<hongkuny@google.com>:

    Creates a benchmark dir

--

PiperOrigin-RevId: 249883771

7f9db598

Transformer v2 benchmark (#6860) · f2ea2f53

Toby Boyd authored May 24, 2019

* Moved common keras code to utils.

* Initial 1 gpu benchmark

- Aligned flags with resnet example
- removed code/features that are not super useful
- eval as part of train if bleu source/ref provided
- add exp_per_second hook

* Rename benchmark classes, pass batch-size and log_steps.

* fix docstring

* Predict done with checkpoints inline

- perfzero baseclass

* steps not epochs with smoother training loop.

* do not initialize history outside loop.

* 5000 between eval not 500

* estimator to keras.

* remove epochs var.

* use range not xrange.

* 200K steps for 1 gpu

* fix global step

f2ea2f53

Add a graph optional_next Reset benchmark. (#6876) · 49eaaaf2
rxsang authored May 24, 2019
```
* Add a graph optional_next Reset benchmark.

* Fix lint error.
```
49eaaaf2
Moved common keras code to utils. (#6859) · 3254cabb
Toby Boyd authored May 24, 2019

3254cabb

Merged commit that fixes transformer's predict and eval. (#6874) · b9cab01b

Tian Lin authored May 24, 2019

* Merged commit includes the following changes:
249776315  by tianlin<tianlin@google.com>:

    Internal change

249763206  by tianlin<tianlin@google.com>:

    For TF 2.0 (related to Beam Search), expand cond dims in tf.where(cond, x, y) to make all parameters broadcastable.

--
249392724  by hongkuny<hongkuny@google.com>:

    Internal change

PiperOrigin-RevId: 249776315

* Merged commit includes the following changes:
249823043  by tianlin<tianlin@google.com>:

    Bring back v2 test for predict and eval.

--

PiperOrigin-RevId: 249823043

b9cab01b

23 May, 2019 6 commits

Add a test enabling get_next_as_optional behavior. (#6862) · 92bad0d2

rxsang authored May 23, 2019

* Add a test enabling get_next_as_optional behavior.

* Remove repeated flag.

* Remove trailing space.

* Make the name shorter.

* Fix lint error.

* Refine the benchmark name.

92bad0d2

Fix non dist strat case. (#6867) · 68650c42
rxsang authored May 23, 2019

68650c42

NCF Keras: Add validation every epoch · abe9e96a

guptapriya authored May 23, 2019

Adding validation every epoch allows us to view the progress during training instead of having to wait until the last eval. Mostly useful for manual runs.

abe9e96a

Change batch size and epochs for NCF benchmarks · e8f97a1d

guptapriya authored May 23, 2019

Current batch size 160000 does not converge to the desired HR. So we decrease to 99k which is known to converge. Tested locally and got to 63.5 at epoch 7. Also decreasing number of epochs as I don't see any improvement after epoch 7-8.

e8f97a1d

Merged commit includes the following changes: (#6863) · f06b5716

Hongjun Choi authored May 22, 2019

249580533  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

249566870  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Set up BERT benchmark test.

--

PiperOrigin-RevId: 249580533

f06b5716

Add enable_get_next_as_optional flag. (#6858) · 272a2baa

rxsang authored May 22, 2019

* Add enable_get_next_as_optional flag.

* Set enable_get_next_as_optional to strategy.

* Add comments to explain the flag.

* Remove trailing whitespace.

* Remove trailing space.

272a2baa

22 May, 2019 6 commits

fix lint issues. (#6855) · 3a97b68c
Toby Boyd authored May 22, 2019

3a97b68c
Merged commit includes the following changes: (#6856) · 85bdf764
saberkun authored May 22, 2019
```
249500988  by hongkuny<hongkuny@google.com>:

    Lints

--

PiperOrigin-RevId: 249500988
```
85bdf764

Add Transformer Big Benchmarks + FP16 for other tests. (#6838) · 23f75313

Toby Boyd authored May 22, 2019

* Add big tests.

* fix super

* Add fp16, increase 8xGPU batch-sizes

* Adding the rest of the fp16 tests.

* Big accuracy test batch_perf_gpu

* fix docstrings

* add _run_and_report

* Edited docstrings

23f75313

Merge Transformer V2 to Github (#6846) · c4f34e58

Tian Lin authored May 22, 2019

* Merged commit includes the following changes:
249218656  by tianlin<tianlin@google.com>:

    Deal with imports, fix a typo and make unit tests fast.

--
249198645  by tianlin<tianlin@google.com>:

    Trivial: Remove one empty line before "import tensorflow"

--
249195490  by tianlin<tianlin@google.com>:

    Initialize Transformer TF V2 Model with Keras subclassing implementation. (Compatible with TF V1)

--
249195008  by tianlin<tianlin@google.com>:

    Internal change

249173564  by hongkuny<hongkuny@google.com>:

    Internal change

249079258  by hongkuny<hongkuny@google.com>:

    Internal change

247691534  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

247533725  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

247509295  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

247311355  by wangtz<wangtz@google.com>:

    Internal change

247303127  by wangtz<wangtz@google.com>:

    Minor changes

--
247277315  by wangtz<wangtz@google.com>:

    Internal change

247143663  by wangtz<wangtz@google.com>:

    Internal change

247129668  by wangtz<wangtz@google.com>:

    Internal change

--
247110547  by wangtz<wangtz@google.com>:

    Internal change

247016722  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

246677582  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

245821839  by shiningsun<shiningsun@google.com>:

    Internal change

245353681  by gjn<gjn@google.com>:

    Internal change

245340898  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

245291749  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

245155641  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

244962751  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

244395199  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

244098932  by shiningsun<shiningsun@google.com>:

    Internal change

244019160  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

242965373  by shiningsun<shiningsun@google.com>:

    Internal change

242930998  by shiningsun<shiningsun@google.com>:

    Internal change

242914617  by shiningsun<shiningsun@google.com>:

    Internal change

242510000  by shiningsun<shiningsun@google.com>:

    Internal change

242183356  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

242049350  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

241749800  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

241663771  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

241416315  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

241054800  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

241028555  by yuefengz<yuefengz@google.com>:

    Internal change

240683076  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

240404735  by yuefengz<yuefengz@google.com>:

    Internal change

239892001  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

239316550  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

238316437  by shiningsun<shiningsun@google.com>:

    Internal change

238251867  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

237876559  by taylorrobie<taylorrobie@google.com>:

    Internal change

236346619  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

236182665  by tayo<tayo@google.com>:

    Internal change

235035787  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

234652747  by wangtz<wangtz@google.com>:

    Internal change

234606543  by wangtz<wangtz@google.com>:

    Internal change

233837502  by shiningsun<shiningsun@google.com>:

    Internal change

232033015  by shiningsun<shiningsun@google.com>:

    Internal change

232028127  by shiningsun<shiningsun@google.com>:

    Internal change

232026986  by shiningsun<shiningsun@google.com>:

    Internal change

232024086  by shiningsun<shiningsun@google.com>:

    Internal change

229839087  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

229794563  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

229204024  by shiningsun<shiningsun@google.com>:

    Internal change

228956426  by shiningsun<shiningsun@google.com>:

    Internal change

228610870  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

228586727  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

228564809  by taylorrobie<taylorrobie@google.com>:

    Internal change

227080093  by shiningsun<shiningsun@google.com>:

    Internal change

227052580  by shiningsun<shiningsun@google.com>:

    Internal change

226217872  by gsundeep<gsundeep@google.com>:

    Internal change

225436264  by shiningsun<shiningsun@google.com>:

    Internal change

222283824  by taylorrobie<taylorrobie@google.com>:

    Internal change

219719958  by taylorrobie<taylorrobie@google.com>:

    Internal change

219330498  by taylorrobie<taylorrobie@google.com>:

    Internal change

219241224  by taylorrobie<taylorrobie@google.com>:

    Internal change

218774474  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

218610966  by taylorrobie<taylorrobie@google.com>:

    Internal change

218576353  by taylorrobie<taylorrobie@google.com>:

    Internal change

217815929  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

217776707  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

217749789  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

217029365  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

216803510  by taylorrobie<taylorrobie@google.com>:

    Internal change

216570979  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

215640939  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

214516790  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

212485310  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

212339556  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

210935187  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

210658133  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

207194426  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

206963562  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

206866123  by taylorrobie<taylorrobie@google.com>:

    Internal change

206631416  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

206180517  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

206170536  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

205252141  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

204191308  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

202519641  by scottzhu<scottzhu@google.com>:

    Internal change

201577585  by scottzhu<scottzhu@google.com>:

    Internal change

201299684  by kathywu<kathywu@google.com>:

    Internal change

199825697  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

199655516  by karmel<karmel@google.com>:

    Internal change

199654555  by karmel<karmel@google.com>:

    Internal change

199653009  by karmel<karmel@google.com>:

    Internal change

199209802  by karmel<karmel@google.com>:

    Internal change

198089630  by karmel<karmel@google.com>:

    Internal change

198064765  by karmel<karmel@google.com>:
    Automated rollback of changelist 197740116.

198060863  by karmel<karmel@google.com>:
    Automated rollback of changelist 197920496.

197920496  by kathywu<kathywu@google.com>:

    Internal change

197841416  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

197740116  by kathywu<kathywu@google.com>:

    Internal change

196608994  by karmel<karmel@google.com>:

    Internal change

196172336  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

195867348  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

195725348  by taylorrobie<taylorrobie@google.com>:

    Internal change

195283704  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

194960508  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

194662698  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

194103064  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

193735742  by saeta<saeta@google.com>:

    Support legacy clusters

--
193581866  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

193567921  by taylorrobie<taylorrobie@google.com>:

    Internal change

193072964  by taylorrobie<taylorrobie@google.com>:

    Internal change

192783651  by scottzhu<scottzhu@google.com>:
    Automated rollback of changelist 192714881.

192714881  by scottzhu<scottzhu@google.com>:
    Automated rollback of changelist 192710755.

192710755  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

192374551  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

192346754  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

192298443  by karmel<karmel@google.com>:

    Internal change

192220576  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

191937900  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

191909212  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

191602380  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

191514106  by scottzhu<scottzhu@google.com>:

    Internal change

191355033  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

191327699  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

190938103  by karmel<karmel@google.com>:

    Internal change

190804388  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

190780502  by karmel<karmel@google.com>:

    Internal change

190479716  by karmel<karmel@google.com>:

    Internal change

190289991  by karmel<karmel@google.com>:

    Internal change

189844661  by scottzhu<scottzhu@google.com>:
    Automated rollback of changelist 189816818.

189816818  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

189796731  by kathywu<kathywu@google.com>:

    Internal change

189639056  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

189628781  by karmel<karmel@google.com>:

    Internal change

189267175  by karmel<karmel@google.com>:

    Internal change

189222334  by karmel<karmel@google.com>:

    Internal change

189096159  by karmel<karmel@google.com>:

    Internal change

189085341  by karmel<karmel@google.com>:

    Internal change

188949700  by karmel<karmel@google.com>:

    Internal change

188780170  by karmel<karmel@google.com>:

    Internal change

188525399  by karmel<karmel@google.com>:

    Internal change

188189681  by karmel<karmel@google.com>:

    Internal change

188076840  by wun<wun@google.com>:

    Internal change

188074937  by wun<wun@google.com>:

    Internal change

187764140  by wun<wun@google.com>:

    Internal change

187663993  by wun<wun@google.com>:

    Internal change

187651035  by karmel<karmel@google.com>:

    Internal change

187415223  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

187047094  by saeta<saeta@google.com>:

    Integrate ClusterResolvers with TPUEstimator.

--
186843813  by karmel<karmel@google.com>:

    Internal change

186843458  by karmel<karmel@google.com>:

    Internal change

186348279  by karmel<karmel@google.com>:

    Internal change

186323602  by ashankar<ashankar@google.com>:

    Internal change

185051452  by wun<wun@google.com>:

    Internal change

185039308  by wun<wun@google.com>:

    Internal change

184934877  by wun<wun@google.com>:

    Internal change

184876413  by wun<wun@google.com>:

    Internal change

184868771  by wun<wun@google.com>:

    Internal change

184751017  by wun<wun@google.com>:

    Internal change

184743769  by wun<wun@google.com>:

    Internal change

183894402  by frankchn<frankchn@google.com>:

    Internal change

183863555  by frankchn<frankchn@google.com>:

    Internal change

183249343  by karmel<karmel@google.com>:

    Internal change

183099733  by wun<wun@google.com>:

    Internal change

182981253  by wun<wun@google.com>:

    Internal change

182977703  by wun<wun@google.com>:

    Internal change

182414330  by ashankar<ashankar@google.com>:

    Internal change

182069924  by wun<wun@google.com>:

    Internal change

181544879  by ashankar<ashankar@google.com>:

    Internal change

180618064  by ashankar<ashankar@google.com>:

    Internal change

180591779  by ashankar<ashankar@google.com>:

    Internal change

179724765  by kathywu<kathywu@google.com>:

    Internal change

179713599  by kathywu<kathywu@google.com>:

    Internal change

179110125  by wun<wun@google.com>:

    Internal change

178793339  by kathywu<kathywu@google.com>:

    Internal change

176581733  by wun<wun@google.com>:

    Internal change

175220569  by wun<wun@google.com>:

    Internal change

175183195  by wun<wun@google.com>:

    Internal change

175171381  by wun<wun@google.com>:

    Internal change

175036833  by wun<wun@google.com>:

    Internal change

175031858  by wun<wun@google.com>:

    Internal change

174891336  by wun<wun@google.com>:

    Internal change

174881248  by kathywu<kathywu@google.com>:

    Internal change

174766157  by wun<wun@google.com>:

    Internal change

173723629  by wun<wun@google.com>:

    Internal change

173579695  by wun<wun@google.com>:

    Internal change

173443901  by wun<wun@google.com>:

    Internal change

173340656  by wun<wun@google.com>:

    Internal change

172954249  by wun<wun@google.com>:

    Internal change

172926654  by wun<wun@google.com>:

    Internal change

172922540  by wun<wun@google.com>:

    Internal change

172921684  by kathywu<kathywu@google.com>:

    Internal change

170943723  by wun<wun@google.com>:

    Internal change

170927898  by wun<wun@google.com>:

    Internal change

170430173  by wun<wun@google.com>:

    Internal change

170428312  by wun<wun@google.com>:

    Internal change

PiperOrigin-RevId: 249218656

* Revert "Merged commit includes the following changes:"

This reverts commit e30d805d3770d7473166dcca1ecb6050dc5de47a.

* Merged commit includes the following changes:

PiperOrigin-RevId: 249232280

* Merged commit includes the following changes:
249376444  by tianlin<tianlin@google.com>:

    Internal change

249373328  by hongkuny<hongkuny@google.com>:

    Clean up tf import

--
249340512  by hongkuny<hongkuny@google.com>:

    Internal change

249325089  by hongkuny<hongkuny@google.com>:

    BERT 2.0

--
249260134  by tianlin<tianlin@google.com>:

    Internal change

PiperOrigin-RevId: 249376444

* Fixed absl_app in official/mnist/mnist_tpu.py

* Make metrics in order (switch from dict to list of tuples).

* Revert ResNet Model to an earlier version.

* Fix lint format.

* Disable two tests in v2.

c4f34e58

Add tweaked fp32 test; Run accuracy tests with LR ops (#6843) · 4726c5b9
Haoyu Zhang authored May 22, 2019

4726c5b9

Merged commit includes the following changes: (#6847) · 30d14a96

saberkun authored May 21, 2019

249377254  by hongkuny<hongkuny@google.com>:

    Internal change

249373328  by hongkuny<hongkuny@google.com>:

    Clean up tf import

--
249333938  by hongkuny<hongkuny@google.com>:

    Fix tf1 import

--
249325089  by hongkuny<hongkuny@google.com>:

    BERT 2.0

--
249173564  by hongkuny<hongkuny@google.com>:

    Internal change

PiperOrigin-RevId: 249377254

30d14a96

21 May, 2019 1 commit
- Replace GlobalAvgPooling with reduce mean to reduce cross-device overhead (#6837) · 1529b82c
  Haoyu Zhang authored May 21, 2019
  
  1529b82c
20 May, 2019 1 commit
- Handle empty eval results in estimator_benchmark. (#6827) · 7ac267a8
  Ayush Dubey authored May 20, 2019
```
* Delete accuracy if exists in eval results.

* get global_step only if it exists in eval results
```
  7ac267a8
18 May, 2019 2 commits
- Include flags when reporting benchmark. (#6809) · bdae51af
  Reed authored May 17, 2019
```
This will allow one to easily reproduce a benchmark by running with the flags.
```
  bdae51af
- Do not return immediately after train_and_evaluate. (#6807) · 23662bb4
  Ayush Dubey authored May 17, 2019
  
  23662bb4
15 May, 2019 3 commits

Fix lint errors. (#6788) · 03027e33
Rachel Lim authored May 15, 2019

03027e33

Set the --clone_model_in_keras_dist_strat to None. (#6781) · 2d4cfad0

Igor authored May 15, 2019

* Set the --clone_model_in_keras_dist_strat to None.  Remove the separate no_cloning benchmarks and add a couple of cloning ones.  Fixes the learning rate schedule to cache its ops per graph.

2d4cfad0

Adds keras imagenet benchmarks which use tf.data's `experimental_slack` option. (#6744) · 6aa6bac5

Rachel Lim authored May 15, 2019

* Added 'tfdata_exp' version of all benchmarks which set
FLAGS.tf_data_experimental_slack = True. Renamed
`data_prefetch_with_slack` to `data_delay_prefetch` (haoyu's change)
to make the names more distinct.

* Add flag to resnet input pipeline and surface through
keras_imagenet_main.py

6aa6bac5

11 May, 2019 2 commits

Remove flacky test: test_bad_seed (#6761) · 03242e38
Toby Boyd authored May 10, 2019
```
- Test passes locally python3 and test is already
    skipped for python2.
```
03242e38

Add FP16 to transformer with benchmark tests. (#6756) · b7e97bec

Toby Boyd authored May 10, 2019

* Add FP16 and benchmarks.

* add missing run and report.

* Add loss_scale as option not included with dtype.

* move loss_scale validation under dtype conditional.

* add loss_scale to flags tested.

b7e97bec

10 May, 2019 5 commits
- Fix trivial model to work properly with fp16 (#6760) · c0b31c51
  Haoyu Zhang authored May 10, 2019
```
* Fix trivial model to work properly with fp16

* Add comment on manual casting
```
  c0b31c51
- Minimize variables and computation in trivial model (#6759) · 5e876e6e
  Haoyu Zhang authored May 10, 2019
```
Previously we had one dense layer in trivial model. The weight was [224*224*3, num_classes].
Using two dense layers, the weights are [224*224*3, 1] and [1, num_classes].
```
  5e876e6e
- Do not report accuracy metrics for benchmark tests (#6757) · cfa37aab
  Haoyu Zhang authored May 10, 2019
```
* Do not report metrics in performance benchmarks

* Rename flag
```
  cfa37aab
- Fix broken test in V2 (#6755) · bae940dc
  Haoyu Zhang authored May 10, 2019
  
  bae940dc
- Use LR schedule ops instead of LR callback for tweaked tests (#6745) · 4b4dbad1
  Haoyu Zhang authored May 09, 2019
```
* Modified tweaked tests to use tensor learning rate
```
  4b4dbad1
09 May, 2019 1 commit

Transformer instrumented for benchmarking (#6734) · 40543869

Toby Boyd authored May 09, 2019

* Add first benchmark and return stats.

* Remove print statements update training steps.

* Revert print T: in print statement.

* Remove print(stats)

* add 2 gpu accuracy test for base.

* Fixed total_batch_size when using gpu + gFile deprecations.

* 8 GPU test name fix

* Add 4 and 8 GPU tests.

* typo fixes.

* Clean up test names and methods.

* bleu uncased.  docstring format fix.

40543869